Indie game storeFree gamesFun gamesHorror games
Game developmentAssetsComics
SalesBundles
Jobs
Tags
A jam submission

GPT-6 Needs ARC EvalsView project page

We propose legislation mandating evaluations of SOTA language models to test for dangerous capabilities.
Submitted by jakub151
Add to collection

Play project

GPT-6 Needs ARC Evals's itch.io page

Results

CriteriaRankScore*Raw Score
Topic#15.0005.000
Generality#54.0004.000
Novelty#84.0004.000
Overall#134.0004.000

Ranked from 1 rating. Score is adjusted from raw score by the median number of ratings per game in the jam.

Judge feedback

Judge feedback is anonymous.

  • Clear and great intervention: Make ARC Eval legally required. It is a great overview of existing approaches to this method along with challenges and opportunities. Some great next steps would be to investigate the feasible implementation of the approach.

What are the full names of your participants?
Carson Ellis, Jakub Kraus, Vidya Silai, Vincent Chung

What is your team name?
The Ablations

Which case is this for?

GPT-6 release considerations

Which jam site are you at?

Online

Leave a comment

Log in with itch.io to leave a comment.

Comments

No one has posted a comment yet